No doubt it'll take a while...
You said you have 36 databases. Could you move half of them using
pg_dump/pg_restore over a few outage windows? (Doing it in bite-sized
pieces reduces risk.)
On 2/21/19 2:27 AM, Julie Nishimura wrote:
Thank you for the suggestions! We realized we cannot add mo
Thank you for the suggestions! We realized we cannot add more space to the
existing cluster due to the hardware limitations. So, we decided to go the
other route by introducing new standby on a new host with bigger size for data
(with pg_basebackup and putting Master into archive mode), then pro
Hello all,
If I were in your situation, I would analyze if it could move only a part
of the 36 databases to the new disk.
* Either, I will move some of the databases to the new disk,
* Either, In the largest databases, I will consider to work in multiple
tablespace configuration, using the command
On 2/19/19 5:02 PM, Julie Nishimura wrote:
Hello, we are almost out of space on our main data directory, and about to
introduce new directory to our cluster. We cannot use multiple physical
disks as a single volume, so we are thinking about creation new tablespace.
Our current data_directory sh
Hello, we are almost out of space on our main data directory, and about to
introduce new directory to our cluster. We cannot use multiple physical disks
as a single volume, so we are thinking about creation new tablespace.
Our current data_directory shows as follows:
/data/postgresql/9.6/main
po
Hello, we are almost out of space on our main data directory, and about to
introduce new directory to our cluster. We cannot use multiple physical disks
as a single volume, so we are thinking about creation new tablespace.
Our current data_directory shows as follows:
/data/postgresql/9.6/main
pos
